A Genomic Analysis of Wolf-Like Canids
Introduction Large dispersal distances within North America have led to an array of admixture zones that four differing wolf-like canids can potentially interbreed; the Great Lakes wolf (C. lycaon or C. lupus lycaon), the coyote (C. latrans) and the red wolf (C. rufus), all originating in North America, along the gray wolf (Canis lupus) of ‘Old World’ derivation. The evolution of these wolf canids has been explained by either commixture of coyotes and varieties of the gray wolf or via a parallel evolution of wolf-like phenotypes arising entirely independently from coyotes (i.e. ‘New World’ evolution) (1). Knowing the origin, whether through ancient hybridization or a ‘New World’ evolution, would allow for advancements in the preservation efforts and legal protections (2–4). High-density single nucleotide polymorphism (SNP) genotyping arrays were used to broaden the current understanding of wolf population history and relations among relatives. Questions about diversification and commixture of traditional wolf-like canids, including the gray wolf, red wolf, Great Lakes wolf, and coyote, were assessed with the SNP-genotyping array (1). These species were specifically studied due to their characteristics of high mobility and weak intraspecific differentiation (2,5–7). A panel of 208 gray wolves from Eurasia and North America, 57 coyotes and 12 red wolves (a total of 48K loci) were assayed in an attempt to more clearly resolve population structure. Contradicting theories concerning the ancestry of the Great Lakes wolf and red wolf were also tested with an analysis of haplotype blocks (1). Genomic Analysis First, the Affymetrix Canine SNP Genome Mapping Array (version 2) was utilized to assess variation in the wolf-like canids. Large and/or populations that were expanding (e.g. Canadian and Western wolves) were found to have high levels of SNP variation (e.g. heterozygosity). The lowest levels of SNP variation were found in captive red and Mexican wolves along with historically constrained populations (e.g. Spain, Italy, red and Mexican wolves). A principle component analysis (PCA) was performed on each individual SNP to distinguish domestic and wolf-like canids followed by Bayesian ancestry inference using the program, structure. From these analyses, it was found that the red and Great Lakes wolves show consistent commixture with coyotes while Midwestern/Southern and Northeastern coyotes show only a small amount of commixture with grey wolves and dogs (Figure 2). In order to assess potential subdivision of populations not commixed, data from all wolf sets were combined and analyzed in structure (‘Old’ and ‘New World’ wolves separately). Neighbor-joining trees were further constructed based on genome-wide allele sharing allowing for the examination of any topological relationships among population groupings. It was noted from all trees that there is a high occurrence at which individuals cluster to their population of origin (1). To assess linkage disequilibrium (LD), which is expected to be greater in inbred and admixed populations (8–11), the physical distance at which the genotypic correlation within the 48K SNP data set falls below a threshold of 0.51. To further distinguish between the effects of inbreeding and admixture on LD, genome-wide autozygosity was examined (via runs of homozygosity (ROH)) to more directly measure recent inbreeding (12). It was observed from this assessment that the Mexican and Isle Royale National Park wolves relative to outbred populations have higher ROH – suggesting both ancient and recent inbreeding. Great Lakes wolves, on the other hand, were observed to have low ROH but a slightly larger LD, indicating recent commixture rather than prior inbreeding. The program SABER was utilized in an attempt to assign ancestry of chromosomal blocks based on two or three reference ancestral populations. It was observed that Great Lakes wolves had, on average, approximately 14.9% of their genome assigned to coyotes. These correlations suggest a distinct genomic composition and ancestry of the two canids. It was also observed that the phenotypically unique Algonquin wolf (presumed as the last population containing a substantial component of the Eastern wolf genome (13), had approximately 58% of its genome derived from gray wolves and the remainder from coyotes. Lastly, the number of generations since initial admixture was measured with SABER in which the number of generations is estimated as a function of the inverse of the inferred chromosomal block size across all assigned blocks (as an average). Between Great Lakes wolves and coyotes it is estimated that 297 +/- 24 generations have arised since initial admixture, and between red wolves and coyotes it is an estimated 144 +/-5 generations (1). Results/Discussion Both the population structure and relationships amongst wolf-like canids were more clearly resolved with the canine SNP genotyping array as it was shown that even within the highly dispersing gray wolf species are regional and continental patterns of genetic subdivision. These genetic segregations were not previously identified in mtDNA sequencing or studies of microsatellite loci (2,6), therefore providing a basis for further studies of wolf-like canid genetic partitioning. It was also observed that Mexican wolves arise as the most genetically unique canid (1), validating the previous hypothesis that an ancient invasion from Eurasia produced this subspecies14. Both the structure and SABER analyses support the hypothesis that red wolves and coyotes are close relatives that diverge somewhat due to a history of commixture with grey wolves. From the data, it is estimated that initial commixture between gray wolves and coyotes was approximately 144 generations ago, implying that a portion of the distinctive red wolf phenotype may be due to historic hybridization of gray wolves and coyotes. Again, both analyses showed genetic distinction between the Great Lakes wolves and Western gray wolves although no association between Great Lakes wolves and red wolves. Distinction reflecting either subspecies, ecotype, or distinct population status, however, is debated (15–17). In studying the admixture in coyotes, it was observed from the structure and SABER analyses that Midwestern/Southern and Northeastern coyotes have a significant amount of admixture with grey wolves and dogs. This result reaffirms the previous notion concerning the importance of relative species abundance in admixture and also suggests that the processes giving rise to the red wolf and coyote might be similar. Overall, a hierarchy of genetic groupings from individual populations to regional and habitat-associated populations was revealed along with a coyote-wolf admixture zone stretching from Southern Texas to the Great Lakes (the largest ever described for a terrestrial vertebrate (18)). This result goes to show and support the great dispersal capability of wolf-like canids and has important implications for conservation policy (1), as current efforts are channeled towards admixed populations whose genomes are as a result of recent habitat changes and/or predator control efforts (19,20). References 1. vonHoldt, B. et al. A genome-wide perspective on the evolutionary history of enigmatic wolf-like canids. Genome Res 21, 1294–1305 (2011). 2. Vila et al. Mitochondrial DNA phylogeography and population history of the grey wolf canis lupus. Mol. Ecol. 8, 2089–2103 (1999). 3. Kyle, C. et al. Genetic nature of eastern wolves: Past, present and future. Conserv Genet. 7, 273–287 (2006). 4. Leonard, J. A. & Wayne, R. K. Native Great Lakes wolves were not restored. Biol. Lett. 4, 95–98 (2008). 5. Wayne, R., Lehmen, N., Allard, M. & Honeycutt, R. Mitochondrial DNA variability of the gray wolf: genetic consequences of population decline and habitat fragmentation. Conserv Biol 6, 559–569 (1992). 6. Roy, M. S., Geffen, E., Smith, D., Ostrander, E. A. & Wayne, R. K. Patterns of differentiation and hybridization in North American wolflike canids, revealed by analysis of microsatellite loci. Mol. Biol. Evol. 11, 553–570 (1994). 7. Forbes, S. & Boyd, D. Genetic structure and migration in native and reintroduced Rocky Mountain wolf populations. Conserv 11, 1126–1234 (1997). 8. Pritchard, J. K. & Przeworski, M. Linkage disequilibrium in humans: models and data. Am. J. Hum. Genet. 69, 1–14 (2001). 9. Tang, H., Coram, M., Wang, P., Zhu, X. & Risch, N. Reconstructing genetic ancestry blocks in admixed individuals. Am. J. Hum. Genet. 79, 1–12 (2006). 10. Gray, M. M. et al. Linkage disequilibrium and demographic history of wild and domestic canids. Genetics 181, 1493–1505 (2009). 11. Gaut, B. S. & Long, A. D. The lowdown on linkage disequilibrium. Plant Cell 15, 1502–1506 (2003). 12. Boyko, A. R. et al. A simple genetic architecture underlies morphological variation in dogs. PLoS Biol. 8, e1000451 (2010). 13. Rutledge, L. Y., Garroway, C. J., Loveless, K. M. & Patterson, B. R. Genetic differentiation of eastern wolves in Algonquin Park despite bridging gene flow between coyotes and grey wolves. Heredity 105, 520–531 (2010). 14. Garcia-Moreno, J., Matocq, M., Roy, M., Geffen, E. & Wayne, R. Relationships and genetic purity of the endangered Mexican wolf based on analysis of microsatellite loci. Conserv Biol 10, 376–389 (1996). 15. Cronin, M. A. & Mech, L. D. Problems with the claim of ecotype and taxon status of the wolf in the Great Lakes region. Mol. Ecol. 18, 4991–4993; discussion 4994–4996 (2009). 16. Koblmüller, S., Nord, M., Wayne, R. K. & Leonard, J. A. Origin and status of the Great Lakes wolf. Mol. Ecol. 18, 2313–2326 (2009). 17. Wheeldon, T. & White, B. N. Genetic analysis of historic western Great Lakes region wolf samples reveals early Canis lupus/lycaon hybridization. Biol. Lett. 5, 101–104 (2009). 18. Barton, N. & Hewitt, G. Analysis of hybrid zones. Annu Rev Ecol Syst 16, 113–148 19. Lehman, N. et al. Introgression of coyote mitochondrial DNA into sympatric North American gray wolf populations. Evolution 45, 104–119 (1991). 20. Wayne, R. & Jenks, S. Mitochondrial DNA analysis implying extensive hybridization of the endangered red wolf Canis rufus. Nature 351, 565–568 (1991).